We have submitted the following draft:
The use of Multiple Locations in the Location Object
draft-morris-geopriv-location-object-issues-00.txt
It is a discussion on the Location Object that, when consolidated,
will probably be part of draft-ietf-geopriv-reqs-00.txt
You may find the draft at:
http://www.ersue.de/geopriv/draft-morris-geopriv-location-object-issues-00.t
xt
Please let me know, if you have any comments.
Regards,
Jorge
-----------------------------------------------------
Internet Draft
John B. Morris, Jr.
Center for Democracy and Technology
J. Cuellar
Siemens AG
A. Gogic
QUALCOMM, Inc.
D. Mulligan
A. Burstein
Samuelson Law, Technology, and Public Policy Clinic
Expires: Dec. 2002 June 2002
The use of Multiple Locations in the Location Object
<draft-morris-geopriv-location-object-issues-00.txt>
Table of Contents
1. Abstract 2
2. Summary 2
3. Conventions Used in This Document 3
4. Underlying Assumptions 3
4.1. Location Representation in the Location Object 4
4.2. Location Representation Format 4
4.3. Provisions for Precision and Confidence 4
4.4. Multiple Representations of a Single Location 5
5. User-controlled Precision of Location Representation 5
6. Misstatement of Location Information 5
7. Multiple Locations 6
7.1. General Principles 6
7.2. The Semantics of Multiple Locations within a Single Object 6
8. Acknowledgements 7
9. References 7
10. Author's Addresses 7
11. Full Copyright Statement 8
1. Abstract
This document discusses three major questions that were posed and discussed
at some length at the interim meeting in San Diego, June 2002:
(1) Should geopriv facilitate the misrepresentation of location information?
(2) Should the geopriv Location Object (LO) accommodate multiple locations
as part of a single positioning transaction?
(3) If so, should the Location Object hold multiple locations in a single
object, or should the multiple locations be contained in multiple objects?
In this paper we propose an answer to those questions.
2. Summary
In this paper we propose the following:
(1) Geopriv should not facilitate the misrepresentation of location
information (but it should also not try to prohibit it).
(2) The protocol should allow multiple Locations within one Location Object,
meaning that the intended location is one of the Locations included in the
LO.
(3) Further each Location may contain different representations of the
location (for instance, the results of different measuring technologies).
(4) An application may use multiple locations contained in multiple objects
if desired.
The relationship between LO, Locations and Location Representations may be
seen schematically as follows: The Location Object MAY contain zero, one, or
several Locations (= Location Fields) L1, L2, etc:
LO =
+------+------+------+------+------+------+--------+------+
| ID | Cred | .. | L1 | .. | Li | .. | Ln |
+------+------+------+------+------+------+--------+------+
\_______________ ___________________/
\/
Location Information
The intended semantics of the Location Information is then that one of the
Li is
Location Information = L1 or L2 or ... Ln
in the sense that the Location Information "holds" (for whatever purpose the
using protocol uses the location) exactly if one location Li "holds".
Further, a Location (field) Li is MAY contain different "Location
Representations"
Li (i=1,.., n) =
+--------+--------+--------+--------+--------+
| LRi1 | LRi2 | LRi3 | .. | LRim |
+--------+--------+--------+--------+--------+
The intended semantics of a Location Li is
Li = LRi1 and LRi2 and ... LRim
in the sense that location Li "holds" (for whatever purpose the using
protocol uses the location) exactly if all location representations
LRij "hold".
--------------------------------------------
Dr. Jorge R Cuellar T +49 89 636-47 585
Security
CT IC 3
Siemens jorge.cuellar@mchp.siemens.de
----------------------------------------------
Received on Tue Jun 25 04:38:33 2002
This archive was generated by hypermail 2.1.8 : Thu Jan 22 2004 - 12:32:23 EST